Browse
→ Code & Dev
→ onUI MCP Server
onUI MCP Server
UI annotation MCP server that lets users annotate and draw on any web UI for AI agents and export structured context without requiring app code changes. Lightweight browser extension for Chrome, Edge and Firefox combined with a local MCP bridge. Provides two capture modes: annotate mode for element-level fixes and draw mode for layout or spacing regions. Four output formats from compact to forensic. All data stays local and never leaves the device. Open source.
MCP unverified
Integration
| Transport | stdio |
| Auth | none |
| Endpoint | https://onui.onllm.dev |
Use Cases
| 01 | Annotate web UI elements and regions directly in the browser through onUI and feed structured context to AI coding agents via MCP for precise element-level fixes without sharing screenshots or writing CSS selectors manually |
| 02 | Enable frontend developers to draw on layout regions and batch-select UI elements through onUI browser extension and have AI agents read fresh annotations via onUI MCP tools for iterative UI pair programming with visual context |
| 03 | Build AI-powered web design review workflows where designers annotate issues on live sites through onUI and AI agents process the structured annotations via MCP with forensic-level detail including element metadata and computed styles |
Tags
ui-annotation browser-extension web-development design frontend local-first developer-tools accessibility
Machine-readable: /api/servers.json
· JSON-LD schema embedded in <head>